Liraglutide Modulates Appetite and Body Weight Through Glucagon-Like Peptide 1 Receptor-Expressing Glutamatergic Neurons

Abstract

Glucagon-like peptide 1 receptor (GLP-1R) agonists are U.S. Food and Drug Administration-approved weight loss drugs. Despite their widespread use, the sites of action through which GLP-1R agonists (GLP1RAs) affect appetite and body weight are still not fully understood. We determined whether GLP-1Rs in either GABAergic or glutamatergic neurons are necessary for the short- and long-term effects of the GLP1RA liraglutide on food intake, visceral illness, body weight, and neural network activation. We found that mice lacking GLP-1Rs in vGAT-expressing GABAergic neurons responded identically to controls in all parameters measured, whereas deletion of GLP-1Rs in vGlut2-expressing glutamatergic neurons eliminated liraglutide-induced weight loss and visceral illness and severely attenuated its effects on feeding. Concomitantly, deletion of GLP-1Rs from glutamatergic neurons completely abolished the neural network activation observed after liraglutide administration. We conclude that liraglutide activates a dispersed but discrete neural network to mediate its physiological effects and that these effects require GLP-1R expression on glutamatergic but not GABAergic neurons.

Figure 1

Characterization of neuronal subtypes activated by liraglutide. A: Breeding schematic with genetic constructs used to generate vGAT-GFP and vGlut2-GFP mice. B and C: Representative images from the CeA (B) and lPBN (C), showing Fos IHC (red) after saline or liraglutide injection in vGAT-GFP (green) and vGlut2-GFP (green) neurons. Blue is DAPI. Scale bars = 100 μm. Regions of interest for quantification are highlighted by a dotted line. D–H: Quantification of Fos+ neurons per section in five regions. The proportion of Fos+ neurons positive for vGAT-GFP or vGlut2-GFP is indicated (n ≥ 6). Data were analyzed by unpaired Student t test between drug treatments. White bars indicate GFP-negative neurons. ***P < 0.001. BLA, basolateral amygdala; CMV, cytomegalovirus; eGFP, enhanced green fluorescent protein; Lira, liraglutide; Sal, saline; scp, superior cerebellar peduncle.

Figure 2

Generation and validation of vGAT^ΔGlp1r and vGlut2^ΔGlp1r mice. A: Breeding schematic with genetic constructs. B: Semiquantitative RT-PCR of Glp1r in the hypothalamus of male mice (n ≥ 13 per group). Data analyzed by one-way ANOVA. ***P < 0.001 by Tukey multiple comparisons test. C: Representative ISH images showing efficient knockdown of Glp1r (green) in cells containing Cre recombinase (red) in the AP. Blue is DAPI. Inset in left panels are shown at higher magnification in the right panels. Scale bars = 50 μm (left panels) and 10 μm (right panels).

Figure 3

Short-term effects of liraglutide in vGAT^ΔGlp1r and vGlut2^ΔGlp1r chow-fed mice. Males and females are displayed separately (A and B) and combined (C). Change in total food intake (A) and body weight (B) 24 h after liraglutide or saline control injection. CTA test results (C) display the taste preference for saccharin after pairing with either saline, liraglutide, or lithium chloride (positive control). n ≥ 10 (A and B) and n = 8 (C). Repeated-measures two-way ANOVA (A and B) or standard two-way ANOVA (C) was performed followed by Sidak (A and B) or Tukey (C) multiple comparisons test. Open symbols indicate saline treatment, partially filled symbols indicate lithium chloride treatment, and closed symbols indicate liraglutide treatment. *P < 0.05, **P < 0.01, ***P < 0.001.

Figure 4

Long-term effects of liraglutide in vGAT^ΔGlp1r and vGlut2^ΔGlp1r high-fat diet–fed male mice. Body weight change from baseline (A) and daily food intake (B) during a 14-day treatment with once daily liraglutide or saline control injection. All mice were fed a 45% high-fat diet 6 weeks before and during the study (n ≥ 6). Data were analyzed by repeated-measures two-way ANOVA followed by Dunnett multiple comparisons test. **P < 0.01, ***P < 0.001 at each time point compared with saline-treated controls of the same genotype.

Figure 5

Neuronal activation in vGAT^ΔGlp1r and vGlut2^ΔGlp1r mice after liraglutide injection. A and B: Representative images from the CeA (A) and lPBN (B) showing Fos IHC (brown) after saline or liraglutide injection. Sections were counterstained with hematoxylin. Scale bars = 100 μm. C–G: Quantification of Fos+ neurons in each of five brain regions (n ≥ 3). Data were analyzed by two-way ANOVA followed by Tukey multiple comparisons test. Open symbols indicate saline treatment, and closed symbols indicate liraglutide treatment. **P < 0.01, ***P < 0.001.

Figure 6

Identification of liraglutide-activated neurons expressing Glp1r. A: Breeding schematic with genetic constructs used to generate Glp1r-GFP mice. B: Representative images from the CeA and AP showing Fos IHC (red) after saline or liraglutide injection in Glp1r-GFP (green) neurons. Blue is DAPI. Scale bars = 100 μm. Regions of interest for quantification are highlighted by a dotted line. C: Quantification of Fos+ neurons per section in each region. In each column, the proportion of Fos+ neurons positive for Glp1r-GFP is indicated in black (n ≥ 6). Data were analyzed by unpaired Student t test between drug treatments. White bars indicate GFP-negative neurons. **P < 0.01, ***P < 0.001. CMV, cytomegalovirus; eGFP, enhanced green fluorescent protein; Lira, liraglutide; Sal, saline.